Problems with PetG

[Alibert]

I don't know where you buy such low quality filament, that you have to dry it bevore printing...

It is independent of the price/quality. I have had spools of cheap AND very expensive european made PETG have this problem straight out of the vaccuum packaging. This is most likely caused by the manufacturing process / feed stock itself. AND I have seen it develop over time when lying open too long in a humid ambient....

[NitroXpress]

You think it is most likely caused by the manufacturing process or feed stock...
Is this not a indicator for quality ?

I have printed a wind turbine with PETG.
And on the roof, is definitely a humid ambient.
If PETG absorbs so much water over the time, it have to crack in the winter when become frosted, or ?

[Alibert]

I think very few manufacturers extensively test the feed stock pellet when they come in, and maintain a rigid control over storage and handling. I have had a particular batch (color) from one supplier have this humidity problem right from the vacuum package, while six other colors (also batches I presume) were perfectly fine. I have also had very expensive high quality european made PETG filamant have this exact same problem.

The fact that you haven't tells that you either purchased from another source or just have had pot luck with your purchases. I go through a lot of ilament and have used up / am using about 20-30 kg of PETG alone in the past 6 months. So far one color from a chinese and all PETG from european source were problematic.

And again, when exposed to humid atmospheres, all polymers will take up water to some extent. How much depends on the type of polymer, the ambient temperature and the ambient absolute humidity expressed in grams of water per cubic meter of air. Forget relative humidity, it is meaningless (unless stated together with the temperature, so youcan calculte the absolute humidity from it) when it comes to gas-solid or gas-liquid equilibria.

I have had spools of PLA, ABS and nylon which were either too water-rich after manufacturing, or developed this issue after some time being exposed to the ambient. Regardless of source. It is just a physical characteristic of the polymer and can't be helped other than air-tight storage and drying when appropiate. If the manufacturer would thoroughly test each and every batch, and safeguard the handling an storage of the feedstock before it is being used that would solve the out-of-the-box problem (which, as a chemical engineer and knowledgeable of how batch manufacturing processes work, I do not think is done in practice). However, the ambient exposure problem remains.

In central heated conditions (say 20C / 35% RH = 6 grams H2O/m3 absolute humidity) I can leave my spools of filament out open in my workshop for weeks. However with hot humid weather(say 35C / 95% RH = 38 grams H2O/m3) the same spools absorb water in a matter of days to the point I have to vacuum dry them to be useful again.

Having said all the above, unless specifically quoted to be european or US made, most 'brand' filaments originate from the same manufacturing sites in China I suppose. If the box does not specifically say 'made in USA' or 'made in Europe' I would bet my money on chinese origins. No problem, the chinese make excellent stuff as my laptop, smartphone, smart TV's, (actually most electronics) etc all can testify.

As to printed PETG being in the weather, frost will not harm it as the water molecules are so dispersed throughout the polymer and can not form ice crystals. When heating however, as in the nozzle, the water turns to steam which is a gas. This boils out of the polymer. Cold is not a problem, heat definitely is. Unless of course there are pockets of water in gaps in the printed object, which will form ice when freezing and thus expand and possibly crack the plastic. But that is a macroscopic issue, not one on a moleculer level.

[Ama-fessional Molder]

Correct me if I'm wrong, but are there not specification sheets available?

It seems to me like the 3d printing industry is overly generic. The materials one particular manufacturer of filament might use could very well be entirely different grades compared to another.

Abs has hundreds, possibly over a thousand unique grades, and the same goes for every other polymer on the market. In my field we use material and grade specific spec sheets with exact Tg, melt points, processing temps, drying requirements, and a full list of mechanical properties.

I'd like to know if this data is available for filament, or if it's just "abs" or "nylon" that you buy and hope for the best.

Drying polymers is one of the things I am very well versed in also, and it looks like, again, this is something the 3d printing world has been left out on. "Humidity controls" are improper for drying and long term storage at elevated temperatures is one of the worst possible things you can do to a resin... yet I see a lot of people doing exactly that.

I want to start a thread on this stuff but there doesn't seem to be a lot of activity here.

[Alibert]

or if it's just "abs" or "nylon" that you buy and hope for the best.

Yep, that pretty much sums it up. No real material spec sheets whatever, mostly boastful claims about how wonderful it is....

"Humidity controls" are improper for drying and long term storage at elevated temperatures is one of the worst possible things you can do to a resin... yet I see a lot of people doing exactly that

I agree that raised temperature is not the best for the filament. It is also my opinion that drying gels are of little use too. I have a vaccuum oven for drying which goes down to 50-100 mbar. At 60-70C that is below boiling point of water.

Do you have experience with respect to vaccuum drying and how it may affect the filament?

[Ama-fessional Molder]

Vacuum drying is almost the ideal process in terms of manufacturing, because it is a rapid process with minimal exposure to heat. To my knowledge the only negative is cost of the units, but they are amazingly fast and often used on high precision processes.

Desiccant gels or beads are actually extremely effective, the problem is you need airflow through the material for it to be effective in a reasonable amount of time. The way a proper polymer dryer works is that cool air is forced through a bed of desiccant, thus removing nearly all moisture from the air. Following this, the air is heated to a set process temperature and forced through the bottom of a hopper full of resin pellets.

The air is cooled off somewhat before returning to the desiccant bed because warm air wants to hold more moisture and transfer less to the desiccant.

You are correct in that simply sticking desiccant gel or beads in the warm chamber is less than ideal. They certainly will absorb moisture, but you are limited by the fixed temperature point.

All that being said, the other primary issue is this:

Overdrying is impossible.*

You cannot dry a resin too much that it causes degradation. I see the word thrown around a lot in 3dprinting and even in my own industry. "Oh, it's just overdried." No, the proper term is oxidation. The material has been oxidized, and in most cases this destroys the properties of that material. This is what happens in the cases of people drying their spools for days or weeks at a time only to have them snap like dry twigs. It is not an issue of being "overdried".

They leave their storage container at an elevated temperature for far too long and the materials actually begin to degrade from interaction with the oxygen in the environment. In theory, one could store them at any temperature below Tg in pure nitrogen and never have an issue, but that's pretty impractical, and not necessary.

The correct solution is limited drying time. For an example, on particular batch of ABS might call for a 170f drying temperature and dry time of 4 hours. After 4 hours at 170f, using a desiccant bed forced air drying system to maintain a very low dewpoint in the air (the "standard" is -40 degrees but some materials work fine with higher), the material should be down below the moisture content required for processing.

Of course in 3d printing, I don't know the exact requirements for drying, but the polymer processing requirements should be identical to injection molding. Thus far the closest thing to proper drying I have ever seen is the filadryer - but even that is ineffective for proper drying and does not get the moisture content low enough for reliable processing.

Drying your material first, then mounting the spool and printing will no doubt give acceptable results, but the issue lies in the following hours during the print. Many materials can absorb moisture from the environment at a rate that would give you wet material within 30 minutes. A specific grade polycarbonate that I use only takes 15 minutes at 50% RH to exceed the minimum moisture content. That's a typical well controlled home environment.

I have some ideas in mind for a solution to the improper drying and storage I am seeing in this field, I just need some time and motivation to develop the drying unit. Once I do however, I will freely share it and my results with everyone. I just don't want to see some manufacturer steal what might turn out to be an effective design and start making money off of all of us. I guess I should look into open source copyrights? I digress.

And again, this is all coming from someone who doesn even own (yet, it's on the way) a 3D printer. I just know how polymers behave during processing conditions. The idea I have right now could very well turn out to be useless or totally impractical.

*Some materials actually do require a specific moisture content during processing, however, you can always put moisture back into a material and run it through the drying process again. These are usually rare cases of highly engineered, custom resins that I believe would be extremely unlikely for you to use in 3d printing.

[Ama-fessional Molder]

TLDR version:


Even if a filament is stored "correctly" and comes from the factory "dry" in a vacuum sealed package, it will still potentially pick up enough moisture to give you inconsistent print quality. Especially if the seal has even a tiny puncture.

In regards to the comment on the other page about how you shouldn't have to dry it straight out of the package, I can tell you that we used to buy pre-dried nylon resin. It gave great results in the first half hour to hour of running, after that it would pick up too much moisture from the atmosphere and result in junk production. Even in a storage/feed unit kept at 10% RH (like so many I see on these forums) it will still pick up moisture beyond acceptable processing levels.

[Alibert]

Really nice to meet a knowledgeable person here!

I will do some calculations about the gel drier system you describe when I have some time. I have worked with saturated salt solution humidity standards in a past project and need to revive some of that knowledge.

As to the vaccum unit, I made a homebrew one using a thick-walled aluminium soup kettle and a 1" thick perspex lid with gasket and taps/guages. Basic, but it works like a charm using an external dual diaphramg pump and a silcone heater mat+control around the kettle. The pump was the most expensive part at about 140 euro's.

I would be very interested in learnign from your experience. Please keep up the dialogue!

[Ama-fessional Molder]

Well, I wouldn't use "gel" desiccant.

http://www.ppe.com/16pdf/0124.pdf

That's the stuff I use in my dryers, but the key is to circulate air through the bed of desiccant instead of just in it's general vicinity. A professional system will use 2 or 3 beds and cycle the air stream among them, regenerating the beds which are not actively drying and thus enabling endless drying capacity - at least until the desiccant breaks down over time.

A dual bed system looks like this:
ZEKS Heatless Desiccant Flow Diagram Flat.jpg

The problem is that these systems are prohibitively expensive for the home user. You can spend several thousand dollars on a small drying unit. My suspicion is that even a high speed 3dprinter will never need more than about a kg/hr throughput, while the industrial ones are typically built for a minimum 4-5 kg/hr.

[Barret]

After view the photos attached,I think you could have a trial with below：
1.Turn the printer temperature to 250C.PETG is better material than PLA,the melt point is similar as abs.
2.set the printing speed slowly.